Global ETD Search

Search theses and dissertations gathered from participating repositories worldwide. Every result links back to the library that holds it. No account is needed.

Results

Showing 1 to 20 of 40 for “"Parallelization strategies"”.

  1. Parallelization strategies for extraction of (co)homological information of digital objects.

    Un objeto digital es un conjunto de n-xels en de una imagen digital n-dimensional. El estudio de la conectividad de estos objetos, interpretados de manera discreta, subdividida o continua, ha sido una custión prioritaria desde los inicios del tratamiento de Imagen Digital. Las herramientas …

    sevilla Repository record for Parallelization strategies for extraction of (co)homological information of digital objects. (opens in a new tab)

  2. Performance and Energy Evaluation of Parallelization Strategies for Network on Chip Communication Architectures: Case Study of Canny Edge Detector Application

    … application is implemented with various software parallelization schemes for better performance as compared to the normal serialized application. Also, to analyze the effectiveness of multicore processors, a comparison among sequential and parallelized coding techniques is performed in this …

    concordia Download PDF for Performance and Energy Evaluation of Parallelization Strategies for Network on Chip Communication Architectures: Case Study of Canny Edge Detector Application (opens in a new tab)

  3. A second order finite element method for the solution of the transonic Euler and Navier-Stokes equations

    … scheme; explicit second order balancing terms; parallelization strategies; and, grid adapting based on a spring analogy.

    concordia Download PDF for A second order finite element method for the solution of the transonic Euler and Navier-Stokes equations (opens in a new tab)

  4. Network-aware device placement search for distributed training

    … distributed training requires not only advanced parallelization strategies but also intelligent placement of model components across heterogeneous computing infrastructures. Existing device placement frameworks often assume simplified, uniform network topologies, leading to suboptimal performance …

    gatech Repository record for Network-aware device placement search for distributed training (opens in a new tab)

  5. Designing Benchmark Suites of Irregular Programs

    … accesses, posing challenges in implementation, parallelization, and optimization. This dissertation explores the systematic design of benchmark suites for this important domain. I extracted key code patterns, styles, and hybridization techniques from existing graph codes, generalized them, and …

    texas-state Repository record for Designing Benchmark Suites of Irregular Programs (opens in a new tab)

  6. Investigation of general-purpose computing on graphics processing units and its application to the finite element analysis of electromagnetic problems

    … onto the granular GPU processors. Efficient parallelization strategies for the finite element matrix assembly on a single GPU and on multiple GPUs are proposed. The parallelization strategies for the finite element matrix solution, in conjunction with parallelizable preconditioners are …

    uiuc Repository record for Investigation of general-purpose computing on graphics processing units and its application to the finite element analysis of electromagnetic problems (opens in a new tab)

  7. Electromagnetic simulations in frequency and time domain using adaptive integral method

    … computational loads and implementing effective parallelization strategies. The methods presented in this thesis include an efficient novel magnetic field integral equation (MFIE) testing scheme for AIM, the interlaced fast Fourier transform (FFT) method for solving general scattering problems, a …

    nus Repository record for Electromagnetic simulations in frequency and time domain using adaptive integral method (opens in a new tab) Download file for Electromagnetic simulations in frequency and time domain using adaptive integral method (opens in a new tab)

  8. A Julia-based parallel FEM solution to the shallow water equations

    … independent data structures, algorithms, and parallelization strategies. The implementation solves the depth-averaged shallow water equations using a Petrov-Galerkin testing method with linear triangular elements and leapfrog time integration, and it supports dynamic wetting and drying …

    manitoba Repository record for A Julia-based parallel FEM solution to the shallow water equations (opens in a new tab)

  9. GVF: BPU Based Vector Fitting

    … variety of electromagnetic modules. Since VF and strategies based on it require many iterations to arrive at optimal number of converged poles, it is highly desired to reduce the computational cost of each VF iteration. This thesis advances the applicability of VF to exploit the emerging massively …

    carleton Repository record for GVF: BPU Based Vector Fitting (opens in a new tab)

  10. Parallelization of the set partitioning method for optimally solving the vehicle routing problem

    … project and will be described in greater detail. Parallelization strategies and considerations will also be discussed. The third step of the Set Partitioning method, which involves integer program-ming, may simply require the use of an efficient parallel integer program solver for speedup in this …

    ut-knoxville Repository record for Parallelization of the set partitioning method for optimally solving the vehicle routing problem (opens in a new tab)

  11. Parallel segmentation of range images on a hypercube-connected distributed computer

    … functions are classified as local or global and parallelization strategies developed for each case.</p>

    utk-theses Repository record for Parallel segmentation of range images on a hypercube-connected distributed computer (opens in a new tab)

  12. HW-SW Methods for Modeling and Optimizing Communication for Scalable Training of Deep Learning Models

    … training is its efficiency for various parallelization strategies. In Chapter 7, we motivate this fact and propose FRED, an interconnect along with its communication implementation algorithms that are flexible for various parallelization strategies of distributed training workloads. We …

    gatech Repository record for HW-SW Methods for Modeling and Optimizing Communication for Scalable Training of Deep Learning Models (opens in a new tab)

  13. Accelerating GNN Inference on Multi-Core Systems

    … (over 100,000 vertices). FGI employs different parallelization strategies optimized for diverse graph structures, maximizing the utilization of multi-level cache hierarchies in multi- core systems. We evaluate a number of GNN models with FGI on a 128-core AMD EPYC system. Compared to …

    wustl Repository record for Accelerating GNN Inference on Multi-Core Systems (opens in a new tab)

  14. Revisiting sparse dynamic programming for the 0/1 Knapsack Problem

    … SKPDP is challenging. We propose two parallelization strategies (fine-grain and coarse-grain) for SKPDP on modern multi-core processors and demonstrate a scalable improvement in the performance.

    colostate Repository record for Revisiting sparse dynamic programming for the 0/1 Knapsack Problem (opens in a new tab)

  15. Towards a Unification of Supercomputing, Molecular Dynamics Simulation and Experimental Neutron and X-ray Scattering Techniques

    … facilities, molecular dynamics code and parallelization strategies, this work also provides insight into the relationship between cellulose crystallinity and cellulose-lignin aggregation by performing multi-million atom simulations. Finally, this work introduces concepts to augment the …

    utk-theses Repository record for Towards a Unification of Supercomputing, Molecular Dynamics Simulation and Experimental Neutron and X-ray Scattering Techniques (opens in a new tab)

  16. Building a foundation for the future of software practices within the multi-core domain

    … management. Current trial and error strategies lack a systematic approach that will scale to growing multi-core and multi-processor environments. In fact, current algorithm and data layout conceptual models applied to design, implementation and pedagogy often conflict with effective …

    uvic Repository record for Building a foundation for the future of software practices within the multi-core domain (opens in a new tab)

  17. Software-Hardware Optimizations for Efficient Collective Communications in Distributed Machine Learning Platforms

    … of distributed ML platforms with arbitrary parallelization strategies and multi-dimensional networks. Second, it presents LIBRA, which enhances the bandwidth utilization of hierarchical collective communication algorithms by optimizing multi-dimensional network topologies via analytical …

    gatech Repository record for Software-Hardware Optimizations for Efficient Collective Communications in Distributed Machine Learning Platforms (opens in a new tab)

  18. High Performance Computing on RISC-V Architectures: Exploring Efficiency, Scalability, and Compiler Optimization for Modern Systems.

    … tasks. Additionally, the study also investigates parallelization strategies, utilizing OpenMP and OpenMPI, to en- hance performance, benchmarked results provide empirical evidence of RISC-V’s computational efficiency, comparing its performance against x86 and ARM-based systems. This research …

    ttu Repository record for High Performance Computing on RISC-V Architectures: Exploring Efficiency, Scalability, and Compiler Optimization for Modern Systems. (opens in a new tab)

  19. Dirbtinės bičių kolonijos algoritmo keliaujančio pirklio uždaviniui spręsti lygiagretinimas /

    … shortest route to food around obstacles. Several parallelization strategies have been considered for this algorithm. Artificial Bee Colony (ABC) algorithm is a newer and much less studied population­-based algorithm. It is decided to try to use one of the parallel versions of the ant colony …

    vilnius Repository record for Dirbtinės bičių kolonijos algoritmo keliaujančio pirklio uždaviniui spręsti lygiagretinimas / (opens in a new tab)

  20. Improving K-Mean Clustering: A Comparative Study of Parallelized Version of Modified K-Mean Algorithm for Clustering of Satellite Images

    … optimizations, combined with effective parallelization strategies, can improve the practicality of K-means clustering for high dimensional satellites image analysis. This work underscores the potential of improving K-means clustering frameworks beyond hardware acceleration alone, …

    sdstate Repository record for Improving K-Mean Clustering: A Comparative Study of Parallelized Version of Modified K-Mean Algorithm for Clustering of Satellite Images (opens in a new tab)

Page 1 of 2